Professional Documents
Culture Documents
1, JANUARY 2003
295
I. INTRODUCTION
N THE PAST 40 years, mankind has ventured into space
using well-established rocket technology involving liquid
fuels and/or solid propellants. This approach has the advantage
for astronauts and fragile payloads that the rocket starts slowly
from the surface of the Earth with its full fuel load, and, as the
fuel is burned off, the altitude and speed increase. In addition to
minimizing the aerodynamic and aerothermal loads, this provides relatively modest accelerationsmaximum values of a
few gees are used for human passengers. Because only a small
fraction of the initial mass reaches orbit, rockets of substantial
size are required to place tens of tons into near-Earth orbit.1
Offsetting these remarkable successes is the very high cost of
burning chemical fuel with a modest efficiency in a rocket engine to get out of the Earths gravitational well. Present estimates are that it costs $20 000 to get one kilogram of material
into orbit. Unless alternatives can be found, it seems likely that
mankinds ventures into space will be limited to a few adventures that can only be undertaken by wealthy nationsthe science-fiction writers dream of colonizing the planets and stars
may be unaffordable.
Manuscript received January 14, 2002. This work was performed in connection with Contract DAAD17-01-D-0001 with the U.S. Army Research Laboratory. The views and conclusions contained in this work are those of the author
and should not be interpreted as presenting the official policies or position, either expressed or implied, of the U.S. Army Research Laboratory or the U.S.
Government unless so designated by other authorized documents. Citation of
manufacturers or trade names does not constitute an official endorsement or
approval of the use thereof. The U.S. Government is authorized to reproduce
and distribute reprints for Government purposes notwithstanding any copyright
notation hereon.
The author is with the Institute for Advanced Technology, The University of
Texas, Austin, TX 78759-5316 USA (e-mail: mcnab@iat.utexas.edu).
Digital Object Identifier 10.1109/TMAG.2002.805923
1Saturn
V was
>3000 tons.
296
297
G. Slingatron
The hula hoop concept for accelerating a particle inside a
hollow ring has been taken to an extreme level with the Slingatron concept [10], [11]. Although no significant velocities
have yet been achieved, the construction of a large spiral system
weighing thousands of tons that would be vibrated at 1.6 Hz to
impart momentum to the launch body has been advocated [12].
At the appropriate instant, the body would be ejected from the
launcher. The accelerating force is imparted to the projectile by
the mechanical push provided by the inclined wall of the accelerating tube on the rear edge of the projectile. It is difficult to
see how a practical projectile can be accelerated without considerable local forces and wear.
H. Lasers
Several concepts for accelerating vehicles into space using a
high-power ground-based laser have been discussed (e.g., [13]).
If the vehicle carries a liquid or solid propellant, a continuous
or pulsed laser can heat and/or ignite the propellant to produce
thrust. The capability of pulsed high-power lasers to accelerate
lightweight flying saucer-like objects up to tens of meters
from the ground has been demonstrated recently. This results
from heating the air layer underneath the lightweight vehicle
plus ablation from the underside by the laser pulse, so that the resulting blast reaction provides the accelerating force. With suitable collimation of the beam and if adaptive optics can minimize
atmospheric distortion, the laser system could operate over long
distances. Scaling up this approach to multikilogram payload
sizes will require lasers having power ratings very much greater
than currently available. Kare [14] has estimated that lasers of
50 000 MW, costing tens of billions of dollars, will be required
for launch to space. This concept seems so far in the future that
it has not been considered further in these studies.
I. Gun Choice
Of the options investigated, only EM railguns seem worthy
of further study for this application. This choice was made on
the basis that:
they have already achieved 7 km/s at small scale, and 9 MJ
at 23 km/s;
significant development is being funded for military applications;
they offer the possibility of achieving the muzzle velocities and energies required;
the potential cost savings seem significant based on our
estimates.
Methods of accelerating large masses in large bore railguns
will need to be developed, and some concepts are suggested
here.
III. RAILGUN LAUNCH-TO-SPACE CONCEPT
A civilian launch-to-space system would be a fixed installation, preferably located at a launch site near the equator and
at high altitude. Since there is no requirement for mobility or
slewing, the launcher can be of a degree of sophistication that is
optimized for this application, rather than the simpler type used
for tactical military missions. The energy demands of a high
Fig. 1.
launch mass make it important to maximize the electrical efficiency of the launcher. This can be achieved by modifying the
approaches used in military railguns so as to:
1) use distributed energy supplies to feed electrical power
into the launcher throughout the launch process, rather
than by feeding all the electricity from the gun breech;
2) augment the magnetic field that permeates the launcher
bore by using pulsed external magnets in their vicinity.
These two aspects are combined in the UTSTAR concept [15]
described below. The following steps illustrate the launch sequence and components.
A. Launch Site
The ideal launch site would be situated on or near the equator
and at the highest possible altitude. The EM launcher will be
fixed and oriented so as to maximize the benefit of the contribution of the Earths rotation to the launch velocity. A convenient arrangement would be on the side of, or embedded into, a
suitable mountain, as shown in Fig. 1 [16]. Having the launch
site at high altitude reduces the aerothermal heating load on the
flight vehicle nosetip: 2000 m above sea level was assumed in
this study. The launch site should also be chosen so that launch
noise will not be an issue and so that items disposed after launch
will not pose a downrange safety hazard.
B. Launcher
Military railguns designed for ordnance applications are not
well matched to the launch-to-space mission. For example, military railguns are usually fed with current from the gun breech.
This is convenient for a barrel of moderate length where slewing
is needed for target acquisition, but it is unsuitable for a very
long launcher because a single current feed point results in high
resistive losses. Ordnance railguns also need to be lightweight
and are therefore designed with a simple two-rail configuration, thereby accepting the efficiency penalty of using only the
self-magnetic field of the propulsive current.
In contrast, the UTSTAR concept [15] uses saddle-back augmenting magnets distributed along the launcher (see Fig. 2) to
increase the magnetic field in the bore of the launcher only in
the vicinity of the launch package. The augmenting magnets
produce a magnetic field similar in strength to that produced
by the main current. This allows the rail current to be reduced
while providing the accelerating force, which is advantageous
for the power supply. Also, by reducing ohmic resistive losses,
it increases the efficiency of the launch process. Nevertheless, a
high current is still required. For one case in this study, it was
298
6.6 MA. For comparison, laboratory railguns have demonstrated muzzle energies of 9 MJ with 3.4 MA [17].
The augmenting magnets are only energized when the launch
package is in the vicinity of the launch package so that electrical energy is expended to create the augmenting field only
when the launch package is nearby, rather than filling the entire
launcher bore with magnetic field. The magnets are energized
sequentially at a rate corresponding to the projectile velocity so
that a traveling wave is created that envelops the armature as
it accelerates along the launcher. This distributed feed arrangement also ensures that minimum magnetic energy is left in the
barrel at projectile launch.5 A further advantage is that the multiple feeds arranged along the launcher will provide a distributed
(staged) power input. This ensures increased efficiency compared with military railguns because the resistive losses associated with current transfer along the rails from the power leads
to the armature will be reduced. This is much more important
for this mission than for most military applications because the
launcher length here is so long that breech fed arrangements
would be prohibitively inefficient [18].
The required current and the launch package acceleration are
strongly dependent on the barrel length. Longer barrels are more
expensive but enable lower currents and accelerations to be used
to achieve a specified muzzle velocity. The extended UTSTAR
launcher concept is shown in Fig. 3. For this study, the acceleration was limited to a modest value (by ordnance standards)
of 2000 gees by using a barrel length of 1600 m. The cost of
this long barrel and its associated infrastructure will be offset by
the easier operating conditions and the reduced need for maintenancealthough the optimum choices depend on a more detailed economic evaluation.
The operating risks will also be reduced compared with ordnance applications by using a modest rail current density of
6 kA/mm. This is 15% of the value used for military barrelsthereby reducing heating and stresses in the barrel.6 No
attempt has been made to optimize the operating conditions and
risks during this study since the necessary database does not
exist.
A high efficiency can be achieved with a long UTSTAR
system, since energy is fed into each stage only when the
projectile is present: 80% was assumed here, and even higher
values may be possible. Thus, for a muzzle energy of 35 GJ
5In a simple breech-fed railgun operating at constant current, the magnetic
energy left in the barrel at projectile exit equals the muzzle kinetic energy, so that
the launch efficiency is limited to a maximum of 50%. In contrast, efficiencies
of 80% to 90% are expected for the UTSTAR launcher in this application.
6Copper rails reach the melting temperature at 44 kA/mm and tungsten at 56
kA/mm.
Fig. 3.
Fig. 4.
299
Fig. 5.
300
TABLE II
UTSTAR LAUNCHER PARAMETERS
G. Docking
The final stage of the process will be to use small thrusters
to rendezvous the payload to a docking station where the payload can be removed and transferred to the recipientsISS or
other. The docking point could be the low end of a space tether
[22]. Once the payload is removed, the disposition of the vehicle
structure needs to be addressed. If the cost of getting into nearEarth orbit can be substantially reduced, the items could be disposed of as trash. However, it would seem better on all counts,
including safety, to make use of the components in orbiteven
as raw materials. After all, having 2000 sets of payload structures, rocket motors, and thrusters as an accessible resource
every year should be of value, provided it can be managed successfully.
IV. SYSTEM OPERATING PARAMETERS
To provide some insight into the system parameters, several
component analyses have been undertaken using simple codes
developed at IAT. These separate codes, which do not yet provide an end-to-end evaluation of a launch-to-space system,13
comprise the following.
1) A calculation that links the muzzle energy requirements
for an augmented launcher with the power input.
2) A calculation that estimates the parameters of a rotating
machine system. This includes an estimate of the machine
sizes and of the prime input power for a specified firing
rate.
3) A calculation that estimates the parameters needed to get
the launch package from the launcher muzzle to payload
in a near-Earth orbit. Aerothermal heating and the mass of
coolant required are calculated, as well as the parameters
of the rocket motor [23].
A short discussion of the results obtained with these codes
follows.
A. Pulsed Power
For this study, it was arbitrarily assumed that 100 pulsed highspeed rotating electrical generators would power the UTSTAR
launcher stages. For an assumed power system delivery efficiency of 90%, the input energy per launch will be 50 GJ.
Although inertial energy storage in high-speed rotors is very
11Less
301
TABLE III
TOTAL ENERGY (GJ) REQUIRED TO LAUNCH 500 TONS OF SUPPLIES INTO ORBIT PER YEAR
analyzed and is not an optimized system solution, but it provides parameters that can be used to assess the needs of the
power supply. For this case, a 1600-m launcher was assumed
that had a low-projectile acceleration. This reduces system risk,
although at the expense of a greater investment in the launcher.
Future optimizations will need to address this issue. However,
the results appear to show that the system parameters fall within
the bounds of acceptability, with a caveat relating to the power
input, as discussed below.
C. Launch to Orbit
Using the code developed by Erengil [23] (which is based on
[2]), a series of calculations was performed to assess the tradeoff
between launch velocity and energy requirements. This takes
into account that as the launch velocity is increased, the energy
input has to increase correspondingly, but the amount of rocket
fuel required to circularize the orbit decreases, so that the payload increases and fewer launches are needed for a given total
mass delivered into space. Details of the calculations are given
in Table III, and the results are shown graphically in Fig. 7. The
results show that launching at about 9 km/s is the minimum energetically, but that launching at a lower velocity of 7.5 km/s
has only 13% energy penalty. Clearly, factors other than energy must ultimately be considered in setting the system parameterschiefly technical feasibility and costbut this calculation gives some indication of one of the issues involved.
V. COST ESTIMATES
If the technical feasibility of the railgun as a launch-to-space
concept can be proven, the next step will be to estimate the financial viability. Unless the railgun has prospects for considerably
Fig. 7.
302
VI. CONCLUSION
TABLE IV
CAPITAL COST ESTIMATE
tions, they are within the same range. Thus, it seems likely that
these costs are likely to be an upper bound to the range of capital investment needed. The total cost is comparable to that for
a new Orbiter vehicle.
Assessing the contribution of this cost to each launch requires
an estimate of the equipment lifetime. The results discussed earlier show that 250 kg of payload may be placed in orbit for
each 1250 kg launched. Thus, getting 500 tons of payload into
orbit each year would require 2000 launches per year.15 A
lifetime of five years has been used as the goal for this equipmentthat is, 10 000 launches. This assumption seems quite
feasible for the rotating machinery, but the launch tube is more
difficult to estimate. No experiments have yet been conducted
that have measured the life of barrel components at this level.
The initial goal for a high-pressure high current ordnance gun
is hundreds of shots, but this barrel is designed for much lower
operating pressures, albeit higher velocities. Clearly, this is an
area where further work is necessary. Nevertheless, using this
assumption for the lifetime between total replacement or major
refurbishment, the contribution of the capital cost of the facility
per kilogram in orbit $1320 M/(10 000 250) $528/kg.
This is an attractively small number compared with the shuttle
cost (2.4%).
The cost of electrical power can easily be estimated. For an
energy input per shot of 50 GJ ( 13.9 MWh) at a typical cost
of 8/kWh, the total cost is about $1100which is negligible
compared with other factors.
In summary, the two cost factors that have been estimated
look attractive. The other two major items need further investigation but seem unlikely to alter these conclusions. However,
embedded in these items are other factors that require careful
evaluation before this concept can be considered practical, including:
1) the choice of a sitepreferably equatorial and at high
altitudethat will be safe;
2) the effect of jettisoned itemssuch as armatureson
down-range safety;
3) the safe disposition (preferably the recycling) of orbit circularization components before or after delivery of the
payload to the customer location;
4) a hazard analysis for an aborted/reduced velocity launch.
15This is eight launches per day for 250 operating days or 5.5 launches per
day for 365 days.
7 km/s
303
vided by the rocket motor and hence can increase the useful
payload. It also has an impact on the optimization of the launch
angle and velocity, and, hence, on the launcher. Jones [22] has
suggested that rendezvous and docking with a tether at a lower
altitude than LEO might permit the rocket motor and launch requirements to be eased.
The disposition of the larger rocket motor and associated
structure after orbit velocity has been achieved requires consideration. One option would be to use the last of the fuel on board
to de-orbit the rocket motor and associated components to burn
up in the atmosphere. On the other hand, having used valuable
energy to get the equipment into orbit, it would be preferable
to collect and utilize these components, provided they can be
collected safely.
F. Cost
A necessary condition to start a development program for this
system is that it must have a competitive cost for putting payload
in orbit. Only a portion of the total cost has been assessed in this
study since we do not have the expertise to accurately estimate
items such as launch costs. The focus has therefore been on the
capital cost of the system and the power costs. The estimated
system cost of $1.3B and a component life of 10 000 launches
without replacement yields a cost of about $530/kg into orbit.
It is important to note that this does not include the cost of the
vehicle itself or operational costs on the Earth or in space, and
these items need to be estimated. We assume that NASA will
develop concepts for collecting and distributing the payload in
orbit.
G. Alternate Missions
This study evaluated direct launch from the surface of the
Earth into LEO. Although several significant technical advances
need to be achieved to accomplish this, the benefits seem attractive. Either as an alternative, or as a strategy for the development
of the technology in steps that have useful off-ramps, it may
be worth considering the development of a moon-based launch
system for which the launch velocity requirements would be
much less stringent (about 2.5 km/s). Using resources available
on the moon, this could serve as the forward base for sending
material to Mars in support of a manned mission.17 A system
of this type would have muzzle velocity requirements similar to
those for a long-range artillery system. It is therefore possible
that a common funding base might be used to construct an experimental test and demonstration facility.
Another alternative would be to locate a system of this type
on the International Space Station with the objective of resupply
for the Mars Mission or for other interplanetary missions.
ACKNOWLEDGMENT
The technical support of Dr. M. Erengil of IAT in providing
the launch-to-orbit calculations for this study is gratefully
acknowledged, as are discussions held with Dr. H. Mark of
The University of Texas at Austin and with experts at the
NASA Marshall Space Flight Center, Huntsville, AL, especially J. Jones, R. Sackheim, and Dr. F. Thio.
17I
304
REFERENCES
[1] L. A. Miller, E. E. Rice, R. W. Earhart, and R. J. Conlon, Preliminary
analysis of space mission applications for electromagnetic launchers,
Battelle Columbus Lab., Final Tech. Rep. to NASA on Contract NAS
3-23 354, Aug. 30, 1984.
[2] M. R. Palmer and A. E. Dabiri, Electromagnetic space launch: A
re-evaluation in light of current technology and launch needs and
feasibility of a near-term demonstration, IEEE Trans. Magn., vol. 25,
pp. 393399, Jan. 1989.
[3] B. Turman, Long range naval coilgun technology, Presentation to
NSWL, Dahlgren, VA, Feb. 21, 2001.
[4] L. Perelmutter et al., Plasma propagation and ignition of propellant in
the chamber of a SPETC gun, IEEE Trans. Magn., vol. 35, pp. 213217,
Jan. 1999.
[5] J. W. Hunter and R. A. Hyde, A light gas gun system for launching
building material into low earth orbit, in AIAA, July 3, 1989, 89-2439(1989) and UCRL Preprint 99 623.
[6] A. Hertzberg, A. P. Bruckner, and D. W. Bogdanoff, Ram accelerator: A
new method of accelerating projectiles to ultrahigh velocities, in AIAA,
vol. 26, 1988, pp. 195203.
[7] C. Knowlen and A. P. Bruckner, Direct space launch using ram accelerator technology, in Space Technology and Applications International
Forum2001, M. S. El-Genk, Ed: American Institute of Physics, Feb.
2001.
[8] P. V. Kryukov, BALSADBallistic system for anti-asteroid defense,
in Abstr. 2nd Int. Workshop RAM Accelerators (RAMAC-II), Seattle,
WA, July 1720, 1995.
[9] D. Wilson and Z. Tan, The blast wave acceleratorFeasibility study,
in Space Technology and Applications International Forum2001, M.
S. El-Genk, Ed: American Institute of Physics, Feb. 2001.
[10] D. A. Tidman, Sling launch of a mass using superconducting levitation, IEEE Trans. Magn., vol. 32, pp. 240247, Jan. 1996.
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]